Value of HRCT of the chest in evaluation of ankylosing spondylitis: correlation between pleuro-parenchymal abnormalities and disease activity

Ankylosing spondylitis (AS) is a chronic inflammatory disorder primarily involving the sacroiliac joints and spine. It is associated with both articular and extra-articular clinical manifestations. Pulmonary involvement is a well-recognized comorbidity of AS, even among patients with early disease. The availability of high-resolution computed tomography (HRCT) has enabled better visualization of the entire lung parenchyma and earlier identification of lung pathologies, ranging from mild to more severe involvement, which were previously missed on X-rays. The aim of the study is to establish the role of HRCT in the detection of pleuro-parenchymal manifestations of AS and to look for correlations between these findings and AS activity. Lung CT scans were done for all our patients and 27 patients (90%) had positive HRCT thoracic findings, while the remaining 3 patients (10%) had a normal HRCT thoracic study. Lower lobe fibrotic changes were the most common finding seen in 20 patients (66.7%) followed by bronchial wall thickening in 13 patients (43.3%) and upper and lower fibrotic changes seen in 9 patients (30%). HRCT thoracic findings were more prominent in late AS (disease duration ≥ 10 years) (13 of 13 patients); while 14 of 17 patients with early AS (disease duration < 10 years) had mildly abnormal HRCT findings. The result of statistical analysis showed that there was negative correlation between thoracic findings and disease activity assessed by BASDAI and this correlation was insignificant (p value = 0.5). HRCT offers an accurate and safe method of assessment of lung disease in patients with AS and without respiratory symptoms. Although there was no correlation between pulmonary findings and AS activity assessed by BASDAI, a positive correlation, albeit insignificant, between pulmonary involvement and disease duration had been found.

of lung pathologies, ranging from mild to more severe involvement, which were previously missed on X-rays [2].
Pulmonary involvement encompasses a wide range of well-defined abnormalities including interstitial lung disease (ILD), upper lobe fibrosis, mycetoma formation as a result of secondary infections, emphysema, bronchiectasis, ventilatory impairment as a result of chest wall restriction and sleep apnea. Non-specific pleuropulmonary changes including, septal thickening, parenchymal micronodules, pleural thickening, parenchymal bands, bronchial wall thickening, and blebs have also been described [3,4].
The aim of the study is to establish the role of HRCT in the detection of pleuro-parenchymal manifestations of AS and to look for correlations between these findings and AS activity.

Methods
This cross sectional study included 30 patients (26 males and 4 females), recruited from the outpatient clinic of Rheumatology and rehabilitation department in our institution from November 2018 to August 2019 according to the following criteria:

Inclusion criteria
Patients diagnosed as ankylosing spondylitis, according to the modified New York criteria [5] as relayed from records of inpatients in Rheumatology department.

Exclusion criteria
Patient with known associated chest diseases and nonimmune mediated pulmonary pathology e.g., old or active tuberculosis.
Patients with history of smoking. All cases were subjected to the following: • Informed written consent was obtained from all patients prior to enrollment. • All data related to patients was collected and included: age, gender, duration of disease, history of smoking and tuberculous infection. • Assessment of disease activity using Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) [5]: The index consists of six questions. The responses were marked on a horizontal line measuring 10 cm (from 0 to 10 cm), where the patient evaluates how he feels in relation to each item in the last week, marking on the scale: if the patient is fine ("very well"), he/she marks zero cm, gradually increasing until "very poor", corresponding to a 10-cm mark.

Protocol for HRCT chest study
All patients underwent scans at 16-MDCT scanner, Somatom, Emotion Siemens. Patients were scanned in supine position in full inspiration. I.V. contrast wasn't administered.
After acquisition, acquired images were transferred to dedicated post-processing workstation where volumetric measurements were performed.
Scans were assessed by three radiologists of 20, 10 and 3 years of experience. They were blinded to the clinical and laboratory data. Results were based on consensus agreement.
HRCT scans were evaluated for presence, distribution, and extent of HRCT abnormalities.
Standard CT criteria for ILD were used to establish ILD diagnosis by presence of its characteristic abnormalities (e.g., subpleural opacities, parenchymal bands, thickening interlobular septa, irregular pleural surface, honeycombing) and determining their extent and distribution pattern.
Multidetector high resolution CT protocols were applied with the patients in the dorsal decubitus position, during maximum inspiration, by using an 16-channel MDCT system. Contiguous axial slices of CT scans were obtained at 5 mm intervals, with 5 mm colimation, 130 Kvp, and an automatic tube modulation at 125 mA. All images were obtained at window levels appropriate for lung parenchyma (window width 1000 HU; window level -700 HU). Sagittal and coronal reconstructed images were performed.

Statistical analysis
Data was entered on the computer using "Microsoft Office Excel Software" program (2010) for windows. Data was then transferred to the Statistical Package of Social Science Software program, version 24 (IBM SPSS Statistics for Windows, Version 24.0. Armonk, NY: IBM Corp.) to be statistically analyzed.
Data presented using range, mean, standard deviation, median, quantitative variables, frequency, and percentage for qualitative ones. Pearson correlation coefficients were calculated to assess the association between disease duration, activity and presence of thoracic findings of AS. P values less than 0.05 were considered statistically significant.

Results
This cross-sectional study was conducted on 30 patients with AS, with 26 (86.7%) males and 4 (13.3%) females. The mean age of the patients was 37, with the youngest being 17 years old and the oldest 59 years old. Mean disease duration was 8.9 y ± 4.9 SD years with a range of from 1 to 20 years. In this study, we divided the patients into two groups according to disease duration of ≥ 10 years (13 patients) and disease duration of less than 10 years (17 patents).
When evaluating the disease activity in AS patients using the BASDAI score (BASDAI ˂ 4 inactive diseases and BASDAI ≥ 4 active diseases), mean score was 3.8 ± 1.4 SD, with 10 patients (33.3%) had high disease activity and 20 patients (66.7%) low disease activity.
Most of our study patients (17 patients) had no pulmonary symptoms. Among those who had symptoms, dyspnea on exertion was the most frequent finding (36.7%), with other less frequent symptoms including mild dyspnea (3.3%) and cough (3.3%).
Lung CT scans were done for all our patients with 27 patients (90%) had a positive HRCT thoracic finding (Table 1), while the remaining 3 patients (10%) had a normal HRCT thoracic study.
Interestingly, thoracic HRCT findings were more prominent in late AS (disease duration ≥ 10 years) (13 of 13 patients); while 14 of 17 patients with early AS (disease duration < 10 years) had abnormal HRCT findings ( Table 1).
The result of statistical analysis showed that there was negative correlation between thoracic findings and disease activity assessed by BASDAI and this correlation was insignificant (p value = 0.5).
Although, HRCT findings were more prominent in the late AS patients and all patients presenting with disease duration ≥ 10 years had findings on HRCT, there was positive correlation between patients with thoracic findings and disease duration = 0.27 but this correlation is insignificant (p value = 0.2).

Discussion
Ankylosing spondylitis is a chronic and inflammatory disease that primarily affects the axial skeleton, the entheses and, less frequently, the peripheral joints. The most prevalent symptom of AS is chronic low back pain, and clinical manifestations usually include limited spinal range of motion, predominantly symmetric bilateral sacroiliitis on radiography, and positive serology for HLA-B27 in 90-95% of cases [6].
Pleuropulmonary involvement was long considered as a rare and late manifestation of AS, because these lesions are usually asymptomatic and undetectable on chest radiographs in the early course of disease. Employing thoracic HRCT imaging, not only the classic lung lesion, but also a wide spectrum of subtle abnormalities, has been recently reported with a higher prevalence in patients with AS [7].
In the current study, majority of our AS patients (56.7%) had no respiratory symptoms. The clinical symptoms of the remaining patients in descending order of frequency were dyspnea on exertion (36.7%), mild dyspnea (3.3%) and cough in (3.3%). This is in accordance with Yuksekkaya et al. [3] who reported that nearly half The HRCT findings in our study, as in the previous studies, would favor an inflammatory process rather than a mechanical etiology for the interstitial disease found in AS patients.
Twenty-seven (90%) of AS patients in our study showed positive HRCT thoracic findings, but this percentage has varied in other studies. Our study result was close to the study done by Yuksekkaya et al. [3] who stated that 95.1% of patients had abnormal findings on CT. Again, Senocak et al. [8], using thoracic HRCT, found abnormalities in 17/20 patients with AS.
On the other hand, our results were not inconsistent with the study done by Rezaei et al. [9] in which HRCT was performed in only 14 (23.3%) patients, for ethical issues. Among these, 8 (57.1%) patients had abnormality in HRCT.
Our study revealed that HRCT findings were more prominent in AS patients with disease duration ≥ 10 years (13 of 13 patients), This was in consistent with Ozdemir et al. [7] in which abnormal HRCT findings was detected in 80% of patients with long disease duration ≥ 10 years, and only in 40% in patients with disease duration of less than 10 years. In contrast to our study, Yuksekkaya et al. [3] concluded that thirty-one patients (96.87%) with early AS and only 7 patients (77.8%) with ≥ 10 years disease duration had abnormal HRCT findings.
Our study revealed that the most frequent HRCT finding was lower lobe fibrotic changes (parenchymal bands) seen in two thirds of patients (66.7%). Hasiloglu et al. [10] stated that the most common findings were parenchymal bands detected in 21 patients (27%). In Sampaio-Barros et al. [11] as in the current study showed that nonspecific parenchymal opacities were the most frequent abnormalities. Of note, upper lobe fibrotic changes were less frequently detected as they were detected in 9 patients (30%) and this not in line with Karakose Okyaltırık et al. [12] who stated that apical fibrosis was the most common HRCT abnormality.
The other frequently detected HRCT findings were lung hyper-lucency and hyperinflation 50%, bronchial thickening 43.3% and pleural thickening 20%. This in line with Souza et al. [13] which showed bronchial wall